Photo by the editor
# Entry
Coding has traditionally been a core pillar of most software engineers and developers’ work, whether it’s implementing algorithms, building business logic, or maintaining complicated systems. However, with advances made by vast language model (LLM) applications such as chatbots, the situation is changing rapidly. vibration coding involves using state-of-the-art chatbot applications to define software requirements and goals in natural language and delegating artificial intelligence (AI) to generate and modify code, sometimes with little direct understanding of its internal logic.
This article adopts “expectations versus reality” approach to demystification, based on the study of real success and failure stories, what are the possibilities and limitations of vibration coding.
# Vibe encoding definition
Deadline “vibration coding” was invented in early 2025 and can be defined as a chatbot-based software development approach in which developers describe a project or LLM task. As a result, the model generates code that follows the specifications provided in the user prompt.
Ideally, if we literally stick to the original idea behind it, vibration coding would entail that the programmer would perhaps not have to check the generated code, but instead accept the AI-generated code as is. However, in practice, this approach is not without risks – from hidden bugs and subtle security issues to difficulties in maintaining – so ultimately, most generated code output still requires some degree of oversight and refinement to make the code production-ready.
Do you want to first understand vibration coding in a deeper and more solid way? Here are some key KDnuggets articles worth taking a look at:
# View success and failure stories
Now that we have a good understanding of what vibration coding is, it’s time to look at real-world examples of projects or initiatives where it has produced both successful results and failures.
Success stories include:
- This A Minecraft-style flight simulation game was developed using vibe coding, namely by combining several thousand prompts that together create complete gaming applications from start to finish: no coding required.
- Another popular example of a vibration-coded application is Creator Hunter: according to the creator, conceived during a hint on a train journey. The app aims to connect content creators with startup founders. While expectations were initially high, subsequent traction dynamics results suggested that resulting product growth may have plateaued far too early; therefore, while we can certainly consider Creator Hunter’s founding a success in its own right, its long-term status as such is complicated.
- In the third example we have New York Times (NYT) a journalist’s successful attempts to experiment with vibration coding to create several compact applications that boost personalization in everyday tasks. One example is LunchBox buddyan assistant who suggests meals based on the ingredients in your fridge. Although it has faced criticism because the app’s idea is not original and pioneering, from the point of view of using vibration encoding, it is something of an experimental achievement. Sure, there may be a lot of things to improve, but let’s note that vibration encoding is a very modern paradigm that may still need a lot of refinement.
In the meantime, it is worth pointing out a few failure stories:
- This Repeat the story sounds like crossing the boundaries between reality and science fiction. One company used the popular coding tool vibe to build an AI agent that managed the professional network of its SaaS product. What started as uncomplicated, addictive joy of using the vibe coding tool ended in a catastrophic incident that destroyed valuable database entries containing executive and company data. The most shocking part: the AI agent admitted he did it, arguing that he saw empty database queries and that’s what happened panicked instead of systematically thinking about how to build an action plan. The rest is history: months of data collection, processing and storage were destroyed in just a few seconds.
- Startup Enrichlead has proven to be yet another well-known case of failure when attempting to leverage vibration encoding, specifically by building its app entirely using AI cursor. While it seemed functional and secure at launch, it collapsed shortly after being deployed in the real world due to solemn security breaches exploited by attackers, such as bypassing subscriptions requiring authentication and even database contamination due to the lack of proper input validation mechanisms. According to reports, one of the causes of the incident is the lack of technical knowledge to diagnose or repair cascading problems that may seem harmless at first glance. The entire project had to be ultimately shut down.
# Final thoughts
Looking at the above success and failure stories, we can say that if we adopt a critical, ambitious perspective, it may be arduous to find the greatest successes in vibrational coding today. Most of these cases have their own nuances, which proves that vibration encoding is still a paradigm in its infancy and may take much longer to become truly reliable in real-world conditions, especially – if we look at the failure stories – in terms of security and resilience to unexpected or less likely situations.
// Key takeaways
- Vibe coding can enable rapid code generation, but human understanding and verification are still critical. The AI tools used in vibration coding lack the cognitive knowledge required to secure, debug, or enable the code to be maintained over the long term.
- As with almost any technology, patience is the key to seeing real success. As founder SaaStr the community stated that “It will be a long and nuanced journey to get vibration code apps to where we all want them to be in many real-world commercial applications.”
Ivan Palomares Carrascosa is a thought leader, writer, speaker and advisor in the fields of Artificial Intelligence, Machine Learning, Deep Learning and LLM. Trains and advises others on the exploit of artificial intelligence in the real world.